Disentangling Controllable Object through Video Prediction Improves Visual Reinforcement Learning

DRL
Re-assign community